A Structural Model for Detecting Communities in Networks

TL;DR

This paper introduces a structural model for community detection in networks, focusing on strategic formation and social influence, addressing computational challenges with simulation and autoregressive methods.

Contribution

It develops a novel theoretical framework for analyzing strategic network formation and social influence, incorporating simulation and multi-modal autoregressive techniques.

Findings

Proposes a static game model for network formation

Addresses computational challenges with simulation methods

Discusses identification of social influence effects

Abstract

The objective of this paper is to identify and analyze the response actions of a set of players embedded in sub-networks in the context of interaction and learning. We characterize strategic network formation as a static game of interactions where players maximize their utility depending on the connections they establish and multiple interdependent actions that permit group-specific parameters of players. It is challenging to apply this type of model to real-life scenarios for two reasons: The computation of the Bayesian Nash Equilibrium is highly demanding and the identification of social influence requires the use of excluded variables that are oftentimes unavailable. Based on the theoretical proposal, we propose a set of simulant equations and discuss the identification of the social interaction effect employing multi-modal network autoregressive.